Debug for https://github.com/pytorch/pytorch/issues/134580

gistfile1.txt

 git log v2.4.0..HEAD --oneline
d74039f7010 (HEAD -> repro-24-134580, origin/repro-24-134580) Repro case for #134580
ccdbe084a9e Skip memory_format tests
49f0d3f1111 Update common_modules.py
630cc4ea8b2 Update test_nn.py
b702a483965 Use newer `toAccumulateType` signature in Normalization.cpp

$ git diff v2.4.0
diff --git a/aten/src/ATen/native/Normalization.cpp b/aten/src/ATen/native/Normalization.cpp
index e9e7c001837..16ada4cead5 100644
--- a/aten/src/ATen/native/Normalization.cpp
+++ b/aten/src/ATen/native/Normalization.cpp
@@ -552,7 +552,7 @@ std::tuple<Tensor, Tensor, Tensor, Tensor, int64_t> _batch_norm_impl_index(
   if (input.sym_numel() == 0) {
     Tensor reserve = at::empty({0}, input.options().dtype(kByte));
     auto options = input.options().dtype(
-        at::toAccumulateType(input.scalar_type(), /*is_cuda=*/input.is_cuda()));
+        at::toAccumulateType(input.scalar_type(), input.device().type()));
     auto save_mean = at::empty_symint(c10::SymIntArrayRef({num_features}), options);
     auto save_invstd = at::empty_symint(c10::SymIntArrayRef({std::move(num_features)}), options);

diff --git a/test/test_nn.py b/test/test_nn.py
index b4283cbbad8..cfc3875538e 100644
--- a/test/test_nn.py
+++ b/test/test_nn.py
@@ -8917,7 +8917,9 @@ class TestNNDeviceType(NNTestCase):
                     else:
                         self.assertEqual(hx.grad, hx_device.grad)

-    def test_BatchNorm_empty(self, device):
+    @dtypesIfMPS(torch.float)
+    @dtypes(torch.double)
+    def test_BatchNorm_empty(self, device, dtype):
         mod = torch.nn.BatchNorm2d(3).to(device)
         inp = torch.randn(0, 3, 2, 2, device=device)
         _test_module_empty_input(self, mod, inp)
diff --git a/torch/testing/_internal/common_device_type.py b/torch/testing/_internal/common_device_type.py
index 07caa0ac3ee..72b357971f1 100644
--- a/torch/testing/_internal/common_device_type.py
+++ b/torch/testing/_internal/common_device_type.py
@@ -1398,6 +1398,10 @@ def expectedFailureXPU(fn):
 def expectedFailureMeta(fn):
     return skipIfTorchDynamo()(expectedFailure('meta')(fn))

+def expectedFailureMPS(fn):
+    return expectedFailure("mps")(fn)
+
+
 def expectedFailureXLA(fn):
     return expectedFailure('xla')(fn)

diff --git a/torch/testing/_internal/common_modules.py b/torch/testing/_internal/common_modules.py
index 5e7e3739695..eafc06ae173 100644
--- a/torch/testing/_internal/common_modules.py
+++ b/torch/testing/_internal/common_modules.py
@@ -15,8 +15,9 @@ from torch.testing._internal.common_cuda import TEST_CUDNN
 from torch.testing._internal.common_dtype import (
     floating_types, floating_and_complex_types_and, get_all_fp_dtypes)
 from torch.testing._internal.common_device_type import (
-    _TestParametrizer, _update_param_kwargs, toleranceOverride, tol,
-    skipCUDAIfCudnnVersionLessThan, skipCUDAIfRocm, precisionOverride, skipMeta, skipMPS, skipCUDAVersionIn)
+    _TestParametrizer, _update_param_kwargs, expectedFailureMPS, toleranceOverride, tol,
+    skipCUDAIfCudnnVersionLessThan, skipCUDAIfRocm, precisionOverride, skipMeta, skipMPS,
+    skipCUDAVersionIn)
 from torch.testing._internal.common_methods_invocations import DecorateInfo
 from torch.testing._internal.common_nn import (
     cosineembeddingloss_reference, cross_entropy_loss_reference, ctcloss_reference,
@@ -3433,9 +3434,6 @@ module_db: List[ModuleInfo] = [
                train_and_eval_differ=True,
                module_inputs_func=module_inputs_torch_nn_BatchNorm1d,
                skips=(
-                   # test fails on MPS backend and is being investigated.
-                   # See https://github.com/pytorch/pytorch/issues/100914
-                   DecorateInfo(skipMPS),
                    # tracking here rather than in the list in test_aotdispatch.py as eval mode passes
                    # RuntimeError: tried to get Double out of SymInt
                    DecorateInfo(
@@ -3454,9 +3452,8 @@ module_db: List[ModuleInfo] = [
                train_and_eval_differ=True,
                module_inputs_func=module_inputs_torch_nn_BatchNorm2d,
                skips=(
-                   # test fails on MPS backend and is being investigated.
-                   # See https://github.com/pytorch/pytorch/issues/100914
-                   DecorateInfo(skipMPS),
+                   # See https://github.com/pytorch/pytorch/issues/134580
+                   # DecorateInfo(expectedFailureMPS, 'TestModule', 'test_memory_format'),
                    # tracking here rather than in the list in test_aotdispatch.py as eval mode passes
                    # RuntimeError: tried to get Double out of SymInt
                    DecorateInfo(

$ python test/test_modules.py -v -k test_memory_format_nn_BatchNorm2d_
test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float32 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float32) ... ok
test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float64 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float64) ... ok
test_memory_format_nn_BatchNorm2d_train_mode_cpu_float32 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_train_mode_cpu_float32) ... ok
test_memory_format_nn_BatchNorm2d_train_mode_cpu_float64 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_train_mode_cpu_float64) ... ok
test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16) ... ok
test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32) ... ok
test_memory_format_nn_BatchNorm2d_train_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16) ... FAIL
test_memory_format_nn_BatchNorm2d_train_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32) ... FAIL

======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_train_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2744, in wrapper
    method(*args, **kwargs)
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 419, in instantiated_test
    result = test(self, **param_kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1364, in wrapper
    fn(*args, **kwargs)
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 129, in test_wrapper
    return test(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 205, in wrapped
    return f(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 775, in test_memory_format
    self.assertEqual(
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3636, in assertEqual
    raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!

Mismatched elements: 216 / 216 (100.0%)
Greatest absolute difference: 1.05078125 at index (1, 0, 1, 4) (up to 1e-05 allowed)
Greatest relative difference: 2576.0 at index (0, 2, 5, 0) (up to 0.001 allowed)

The failure occurred for item [0]

To execute this test, run the following from the base repo dir:
     python test/test_modules.py -k TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_train_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2744, in wrapper
    method(*args, **kwargs)
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 419, in instantiated_test
    result = test(self, **param_kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1364, in wrapper
    fn(*args, **kwargs)
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 129, in test_wrapper
    return test(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 205, in wrapped
    return f(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 775, in test_memory_format
    self.assertEqual(
  File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3636, in assertEqual
    raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!

Mismatched elements: 216 / 216 (100.0%)
Greatest absolute difference: 1.0258822441101074 at index (1, 1, 4, 4) (up to 1e-05 allowed)
Greatest relative difference: 1735.79296875 at index (0, 1, 0, 0) (up to 1e-05 allowed)

The failure occurred for item [0]

To execute this test, run the following from the base repo dir:
     python test/test_modules.py -k TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

----------------------------------------------------------------------
Ran 8 tests in 0.362s

FAILED (failures=2)